A note on the Neyman-Johnson technique

A statistical problem which frequently arises in educational and psychological experimentation is that of testing the significance of the difference of the mean scores of two groups on some criterion variable, where the differential effects of one or more variables which are correlated with the criterion must be statistically eliminated. The usual analytical technique for this type of problem is the analysis of covariance (9). The Neyman-Johnson technique (7) provides another, and substantially different, approach. A computational procedure is suggested here which utilizes the advantages of both techniques without an undue increase in computational labor. In addition, the Neyman-Johnson technique is generalized to the case ofn predictor variables. Its application has heretofore been limited to a maximum of three predictor variables.This paper was written while the author was a Psychometric Fellow of the Educational Testing Service, Princeton, New Jersey.     

How you can help,without making a difference

There are many cases in which people collectively cause some morally significant outcome (such as a harmful or beneficial outcome) but no individual act seems to make a difference. The problem in such cases is that it seems each person can argue, ‘it makes no difference whether or not I do X, so I have no reason to do it.’ The challenge is to say where this argument goes wrong. My approach begins from the observation that underlying the problem and motivating the typical responses to it is a standard, intuitive assumption. The assumption is that if an act will not make a difference with respect to an outcome, then it cannot play a significant, non-superfluous role in bringing that outcome about. In other words, helping to bring about an outcome requires making a difference. I argue that the key to solving the problem is to reject this assumption. I develop an account of what it is to help to bring about an outcome, where this does not require making a difference, and I use this explain our reasons for action in the problem cases. This account also yields an error theory that explains why the standard assumption is so tempting, even though it is mistaken.     

Effects of Group Training in Problem‐Solving Style on Future Problem‐Solving Performance

Seventy‐five participants from one suburban high school formed 21 teams with 3–4 members each for the Future Problem Solving Program International (FPSPI). Students were selected to participate in either the regular FPSPI or an enhanced FPSPI, where multiple group training activities grounded in problem‐solving style were incorporated into a 9‐week treatment period. An ANCOVA procedure was used to examine the difference in team responses to a creative problem‐solving scenario for members of each group, after accounting for initial differences in creative problem‐solving performance, years of experience in FPSPI, and creative thinking related to fluency, flexibility, and originality. The ANCOVA resulted in a significant difference in problem‐solving performance in favor of students in the treatment group (F(1, 57) = 8.21, p = .006, partial eta squared = .126, medium), while there were no significant differences in years of experience or creativity scores. This result led researchers to conclude that students in both groups had equivalent creative ability and that participation in the group activities emphasizing problem‐solving style significantly contributed to creative performance.     

Testing differences between nested covariance structure models: Power analysis and null hypotheses

For comparing nested covariance structure models, the standard procedure is the likelihood ratio test of the difference in fit, where the null hypothesis is that the models fit identically in the population. A procedure for determining statistical power of this test is presented where effect size is based on a specified difference in overall fit of the models. A modification of the standard null hypothesis of zero difference in fit is proposed allowing for testing an interval hypothesis that the difference in fit between models is small, rather than zero. These developments are combined yielding a procedure for estimating power of a test of a null hypothesis of small difference in fit versus an alternative hypothesis of larger difference.     

谈“酒精滥用”的诊断意义和伦理学问题

随着经济的发展,酒精使用障碍已经成为一个重大的公共卫生问题,对我国居民酒精使用现状的调查迫在眉睫。在流行病学调查中,因我国文化的差异,DSM-IV酒精滥用诊断标准在实际应用时产生分歧,本文举例说明分歧所在,并阐明文化因素在不同诊断系统中的重要影响以及伦理学问题。     

Basic statistics and the inconsistency of multiple comparison procedures.

This paper has two main themes. First, the various statistical measures used in this journal are summarized and their interrelationships described by way of a flow chart. These are the pooled standard deviation, the pooled variance or mean square error (MSE), the standard error of each treatment mean (SEM) and of the difference between two treatment means (SED), and the least difference between two means which is significant at (e.g.) the 5% level of significance (LSD(5%)). The last three measures can be displayed as vertical bars in graphs, and the relationship between the lengths of these bars is graphically illustrated. It is suggested that the LSD is the most useful of these three measures. Second, when the experimenter has no prior hypotheses to be tested using analysis of variance "contrasts," a multiple comparison procedure (MCP) that examines all pair-wise differences between treatment means, may be appropriate. In this paper a fictitious experimental data set is used to compare several well-known McPs by focussing on a particular operating characteristic, the consistency of the results between an overall analysis of all treatments and an analysis of a subset of the experimental treatments. The procedure that behaves best according to this criterion is the unrestricted least significant difference (LSD) procedure. The unrestricted LSD is therefore recommended with the proviso that it be used as a method of generating hypotheses to be tested in subsequent experimentation, not as a method that attempts to simultaneously formulate and test hypotheses.     

Age Differences in Performance on Two Versions of the Brown-Peterson Task

The reports regarding whether normal aging is associated with faster forgetting in the Brown-Peterson task have been conflicting. We hypothesized that, in light of documented age differences on other tasks involving secondary memory, older adults would show disproportionate forgetting on the Brown-Peterson task as retention interval lengthens. Previous negative results might be a function of the specific experimental procedure used. Experiment 1, using a commonly employed procedure, did not indicate an age-related increase in rate of forgetting. This procedure allowed for differences in rehearsal opportunity, task difficulty, and amount of information to be processed. Experiment 2 controlled for these factors and did reveal significant age differences in the forgetting function. This age difference occurred only at the point where recall became dependent upon secondary memory. There was, however, no evidence of an age-related increase in rate of forgetting from primary memory in either experiment. These findings have implications for theories of cognitive aging as well as for the use and interpretation of a commonly used version of the Brown-Peterson task.     

Alternative-based thresholding with application to presurgical fMRI

Functional magnetic reasonance imaging (fMRI) plays an important role in pre-surgical planning for patients with resectable brain lesions such as tumors. With appropriately designed tasks, the results of fMRI studies can guide resection, thereby preserving vital brain tissue. The mass univariate approach to fMRI data analysis consists of performing a statistical test in each voxel, which is used to classify voxels as either active or inactive—that is, related, or not, to the task of interest. In cognitive neuroscience, the focus is on controlling the rate of false positives while accounting for the severe multiple testing problem of searching the brain for activations. However, stringent control of false positives is accompanied by a risk of false negatives, which can be detrimental, particularly in clinical settings where false negatives may lead to surgical resection of vital brain tissue. Consequently, for clinical applications, we argue for a testing procedure with a stronger focus on preventing false negatives. We present a thresholding procedure that incorporates information on false positives and false negatives. We combine two measures of significance for each voxel: a classical p-value, which reflects evidence against the null hypothesis of no activation, and an alternative p-value, which reflects evidence against activation of a prespecified size. This results in a layered statistical map for the brain. One layer marks voxels exhibiting strong evidence against the traditional null hypothesis, while a second layer marks voxels where activation cannot be confidently excluded. The third layer marks voxels where the presence of activation can be rejected.     

How to solve a mechanical problem: the relevance of visible and unobservable functionality for kea

Animals sometimes succeed quickly in solving a mechanical problem that is a modification of one they have previously learnt to solve. However, they may do so by attending to the visible features of the relevant physical dimension without knowing its causal functionality, if that is not directly perceivable. This kind of problem solving can be tested by simultaneously offering two mechanical devices with the same visual features but different inherent appropriateness for problem solving. Here, we provide data collected by following this procedure for the first time in a bird species. Captive kea, Nestor notabilis, a parrot species highly interested in the affordances of objects, were offered a mechanical problem in which they had to remove a baited tube from one of two upright poles where removal was blocked at the end of one pole but not the other. With extended but not with restricted exploration of a baseline apparatus, the kea immediately succeeded in removing the tube from an apparatus that had modified pole ends when they were able to visually observe (without touching) that one of these ends would block tube removal but the other would not. However, when the kea were allowed to explore two poles that had a removable and a fixed obstruction where the difference in function was not visible, they preferred the removable one during unbaited exploration but failed afterwards to push a tube to the end of the pole with the loose structure during subsequent baited test trials. Thus, in spite of the speed with which the kea learnt the tasks, there was no indication that they understood the underlying unobservable causal structure of the problem.     

The true distributions of the range of rank totals in the two-way classification

This paper presents the results from an investigation of the true probability distributions of the range of rank totals. A procedure for generating an approximation to the true distributions is also given. A comparison of the results of this approximation with an extensive criterion of generated true and sample distributions, and with other approximations is indicated. Accurate estimates of the critical ranges necessary to reach significance at three commonly used alpha levels, where the number of judges and items is less than or equal to sixteen, are presented in tabular form.     

Momo and Ibrahim—‘Thus-and-Otherwise’: A Dialogical Approach to Religious Difference

Studies on religious difference usually take their point of departure in an intellectually biased view of interreligious encounters where difference is treated as a problem to be solved and the challenge of facing the religious other is an intellectual task of making him/her intelligible. The current article aims at providing an alternative to this logical approach to religious difference by outlining a dialogical one. The perspective is based on the works of several Jewish, Christian, and Muslim philosophers of religion who exchange the question of who has the truth for more ethically and practically oriented questions; for example, finding dignity in difference and seeing the religious other as a legitimate perspective on the world. The novel Monsieur Ibrahim and the Flowers of the Koran is used as an empirical arena for addressing the main issue at stake: can a monotheist ever truly appreciate religious diversity?     

SIMPLIFIED HABIT REVERSAL TREATMENT FOR CHRONIC HAIR PULLING IN THREE ADOLESCENTS: A CLINICAL REPLICATION WITH DIRECT OBSERVATION

Three developmentally normal adolescents with chronic hair pulling were treated with a simplified habit reversal procedure consisting of awareness training, competing response training, and social support. Treatment resulted in an immediate reduction to near-zero levels of hair pulling, with one to three booster sessions required to maintain these levels. The results were maintained from 18 to 27 weeks posttreatment, although 1 participant reported difficulty at follow-up. The effectiveness of simplified habit reversal and suggestions for future research are discussed.     

Statistical problems in the evaluation of army tests

The introduction of psychological tests for personnel selection in the British forces has given rise to several novel problems in statistical procedure. The solutions proposed are in the main extensions of devices already familiar in educational psychology. The more important are: (i) where the criterion yields a threefold classification only, a method of triserial correlation or of biserial correlation assuming point-distributions for the extremes; (ii) where the data on which validation has to be based are drawn from a selected sample, a simplified form of Pearson's equations to correct for selection; (iii) where the best line of demarcation has to be deduced from theoretical rather than practical considerations, a formula based on the principle of minimal discrepancy.     

The New Problem of Numbers in Morality

Discussion of the “problem of numbers” in morality has focused almost exclusively on the moral significance of numbers in whom-to-rescue cases: when you can save either of two groups of people, but not both, does the number of people in each group matter morally? I suggest that insufficient attention has been paid to the moral significance of numbers in other types of case. According to common-sense morality, numbers make a difference in cases, like the famous Trolley Case, where we must choose whether to kill a person (or persons) as a side effect of saving a greater number. I argue that recognition of the role of numbers in killing cases forces us to reassess purported solutions to the problem of numbers.     

Regression weights as a function of test length

A battery of pencil-and-paper tests is commonly used for predicting a single criterion. If the score on each test is the number of correct answers, the composite battery score would normally be the sum of the weighted test scores, where the weights are the raw score regression weights. Knowing the reliability of each test, it is possible to alter the lengths of the tests in a manner such that the weights will all be equal. The composite battery score would then simply be the total number of items answered correctly and scoring would be greatly simplified. Such simplification is particularly desirable where the volume of testing is large. Section I of the article outlines the procedure for altering the lengths of the tests, and Section II gives a proof of the method.     

To Be Young,Gifted, and Strategic: Advantages for Memory Performance

Gifted and nongifted children's use of an organizational strategy was contrasted on multitrial free-recall tasks, using different sets of items on each trial. In an initial experiment, gifted children initially had higher levels of recall and strategic functioning than nongifted children, but this advantage was lost on later trials. While overall there was an advantage to memory of being strategic, this advantage was statistically significant for the gifted children only at trial 1, whereas it was significant for the nongifted children on trials 2 through 5. A sort-recall procedure was used in Experiment 2, with results indicating that gifted children benefited more than nongifted children when strategy use was simplified, while the results of Experiment 3, which used nonsense words as stimuli, demonstrated that gifted children demonstrated greater use of active strategies than nongifted children. The results of these experiments were interpreted as evidence that at least a portion of gifted children's advantage on free recall tasks lies in nonstrategic processes.     

A sensible formulation of the significance test

The conventional procedure for null hypothesis significance testing has long been the target of appropriate criticism. A more reasonable alternative is proposed, one that not only avoids the unrealistic postulation of a null hypothesis but also, for a given parametric difference and a given error probability, is more likely to report the detection of that difference.     

SAD as a mathematical assistant—how should we go from here to there?

The System for Automated Deduction (SAD) is developed in the framework of the Evidence Algorithm research project and is intended for automated processing of mathematical texts. The SAD system works on three levels of reasoning: (a) the level of text presentation where proofs are written in a formal natural-like language for subsequent verification; (b) the level of foreground reasoning where a particular theorem proving problem is simplified and decomposed; (c) the level of background deduction where exhaustive combinatorial inference search in classical first-order logic is applied to prove end subgoals.

We present an overview of SAD describing the ideas behind the project, the system's design, and the process of problem formalization in the fashion of SAD. We show that the choice of classical first-order logic as the background logic of SAD is not too restrictive. For example, we can handle binders like Σ or lim without resort to second order or to a full-powered set theory. We illustrate our approach with a series of examples, in particular, with the classical problem .     

The effects of errors on children's performance on a circle-ellipse discrimination

Children first learned by means of a teaching program to discriminate a circle from relatively flat ellipses. Children in the control group then proceeded into a program which gradually reduced the difference between the circle and the ellipses. They advanced to a finer discrimination when they made a correct choice, and reversed to an easier discrimination after making errors ("backup" procedure). The children made relatively few errors until they approached the region of their difference threshold (empirically determined under the conditions described). When they could no longer discriminate the forms, they learned other bases for responding that could be classified as specifiable error patterns. Children in the experimental group, having learned the preliminary circle-ellipse discrimination, were started at the upper end of the ellipse series, where it was impossible for them to discriminate the forms. The backup procedure returned them to an easier discrimination after they made errors. They made many errors and reversed down through the ellipse series. Eventually, most of the children reached a point in the ellipse series where they abandoned their systematic errors and began to make correct first choices; then they advanced upward through the program. All of the children advanced to ellipse sizes that were much larger than the ellipse size at the point of their furthest descent.     

A note on preliminary tests of equality of variances

Preliminary tests of equality of variances used before a test of location are no longer widely recommended by statisticians, although they persist in some textbooks and software packages. The present study extends the findings of previous studies and provides further reasons for discontinuing the use of preliminary tests. The study found Type I error rates of a two‐stage procedure, consisting of a preliminary Levene test on samples of different sizes with unequal variances, followed by either a Student pooled‐variances t test or a Welch separate‐variances t test. Simulations disclosed that the twostage procedure fails to protect the significance level and usually makes the situation worse. Earlier studies have shown that preliminary tests often adversely affect the size of the test, and also that the Welch test is superior to the t test when variances are unequal. The present simulations reveal that changes in Type I error rates are greater when sample sizes are smaller, when the difference in variances is slight rather than extreme, and when the significance level is more stringent. Furthermore, the validity of the Welch test deteriorates if it is used only on those occasions where a preliminary test indicates it is needed. Optimum protection is assured by using a separate‐variances test unconditionally whenever sample sizes are unequal.